The overall goal of this research project is to develop a noninvasive Magnetic Resonance/Optical Imaging test to assess and predict ischemic brain damage and recovery after acute hemorrhage using a rat model. Patients who suffer acute blood loss (e.g. from trauma, surgery, gastrointestinal bleeding) are often initially resuscitated with asanguinous salt solutions before transfusion of blood is available and deemed necessary. Loss of red blood cell mass and hemodilution reduce oxygen delivery to tissues. The brain is particularly sensitive to oxygen delivery;even relatively short periods of tissue hypoxia can result in permanent damage or stroke. Presently, there is no clinically available method to assess adequate delivery of oxygen to the brain and predict stroke;resuscitation and transfusion is guided entirely by arbitrary clinical guidelines. Assessment of adequate oxygen delivery in such patients would lead to more effective treatment. Furthermore, there is no animal model of hemorrhage and resuscitation either to permit evaluation of imaging methods to predict stroke or appropriate resuscitation with asanguinous fluids, blood or oxygen carriers. The overall hypothesis of this proposal is that we can use information provided by magnetic resonance (MR) and diffuse optical tomography (DOT) to predict the status of brain tissue 24 hours after acute hemorrhage and resuscitation. The specific aims are: AIM 1.) Develop a rodent acute hemorrhage/resuscitation model to include recovery and follow-up imaging at 24 hrs that produces ischemic brain damage while optimizes transfusion protocol to avoid mortality. AIM 2.) Measure changes in MR relaxation rates, spectroscopy, and diffusion parameters throughout brain, as well as localized changes in oxy/deoxy/total hemoglobin with DOT during hemorrhage/resuscitation and recovery. These acute measurements will be compared to imaging and histological analysis of stroke 24 hours after hemorrhage. AIM 3.) Evaluate several approaches to characterize the multiparametric measurements during hemorrhage/resuscitation and follow-up time points to predict outcome. We will begin by using ISODATA clustering algorithm and GLM as approaches to predictive mapping. Using the optimum analysis method, we will test the primary hypothesis. A successful outcome from this study would demonstrate that acute noninvasive imaging could be used to define the extent of reversible or irreversible ischemic changes in brain from acute blood loss, as well as the impact and optimal timing of resuscitation with asanguinous fluids, blood or oxygen carriers. Such data would strongly support use of MR and/or DOT to monitor, assess and clinically manage hemorrhaging patients and would point the way for future studies in humans. The overall goal of this research project is to develop a noninvasive magnetic resonance/optical imaging test to assess and predict brain damage and recovery after acute blood loss using an experimental animal (rat) model. A successful outcome from this study would demonstrate that acute diagnostic imaging could be used to identify and define the extent of reversible or irreversible ischemic changes in brain caused by acute hemorrhage, as well as elucidate the impact and optimal timing of resuscitation with blood or artificial blood substitutes. These data would strongly support the application of this technology to assess and manage patients undergoing hemorrhage and would point the way for future studies in humans.